ietf-nntp NEWNEWS deprecation.
dave at wildbear.on.ca
Thu Dec 4 10:34:08 PST 1997
At 09:01 AM 12/4/97 +0100, you wrote:
>- NEWNEWS can get Message-Ids from multiple newsgroups at the same time.
>- NEWNEWS need to store only one mark per server rather than one per server
> per group
These two parameters are very sweet from the perspective of a microcontroller
>If you are not just fetching news for further processing, but fetching
>them for reading, as you "news-in-a-watch" analogy indicates, then NEWNEWS
>is totally useless, because to READ news, you need a lot more than getting
>articles . . .
Sorry, the interpreted context for the use of news in the metaphorical
internet-in-watch is incorrect. I am not suggesting reading news on a
This group and hence this mailing list deal with single largest interaction
NNTP, the internet news pool of newsgroups. And this group should continue to
use the internet news as the base model for implementation and optimization of
However, there are a few other pools of NNTP transaction that are NOT part of
the internet pool and many other communications that could easily be
up-graded to localized NNTP transactions.
Some might argue that such pools are not valid, not of importance, not correct
because they are not connected to the mainstream. Such an arguement becomes
obviously incorrect when one applies the same arguement to a localized IP
network that's not connected to the internet.
Therefore we should all remember that there are other places where NNTP is
Such implementations include insolated (and thereby falsely assumed secure)
NNTP pools for R&D. Such implementations include companies that run local
newsgroups, sales, accounting, marketing, playtime, and public announcements.
Included here should also be systems for communicating with/notifying field
staff (sales, tech support, courier/delivery). With wireless networks reaching
new capabilities and becoming less propritary isolated NNTP pools will
become more common place.
Other than server to server communications the only NNTP pools I know
of are solutions for human communications. However, with the trend to merging,
bridging and reusing protocals I suggest that NNTP will become a common
ground for simple/low-speed inter-microcontroler messaging.
NEWNEWS is only a serious problem for poor implementations with large
amounts of messages and large numbers of unsubscribed newsgroups and
even then NEWNEWS has its use, as others point out.
The draft should re-enforce the importance of implementation and
support for NEWNEWS. It should also discourage any disabling techniques.
A standard that is not standardly available/implemented is not, by its
very nature a standard.
dfielder at sentex.net
WildBear Consulting Inc.
dave at wildbear.on.ca
More information about the ietf-nntp